Service Replication Policy for High Availability Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional server cluster management schemes fail to meet increasing demands for high performance and reliability, particularly in large clusters, leading to prolonged downtime during node failures which can violate service guarantees for high availability services.
Innovation Solution
Implementing a replication policy with configurable parameters such as deployment, prepared, and service replication factors, using a consistent hashing algorithm to quickly identify and switch to standby nodes, reducing downtime by pre-configuring and replicating services across multiple nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional failover policies are used with a single back-up node, then device complexity is reduced, but node startup time during failover increases leading to prolonged downtime
Solution Approach 1:
The patent applies preliminary action by pre-configuring services on multiple standby nodes before failover occurs. When a node fails, the system can immediately activate a pre-prepared service instance on a standby node, eliminating the need to perform configuration and startup procedures during the failover event. This reduces node startup time during failover while maintaining manageable complexity through automated policies.
Solution Approach 2:
The patent segments the failover process by dividing standby nodes into different readiness levels (e.g., fully prepared, partially prepared, and unprepared). This segmentation allows the system to quickly identify and activate nodes with appropriate service instances already configured, reducing failover time without requiring all standby nodes to be fully configured, thus balancing complexity and performance.
2Reliability
If services are replicated across multiple nodes with pre-configuration, then reliability during failover is improved, but device complexity and resource consumption increase
Solution Approach 1:
The patent implements dynamic replication policies where the number and configuration level of standby nodes can be adjusted based on service criticality, historical failure patterns, and current system load. This dynamic approach allows the system to maintain high reliability for critical services while reducing complexity and resource consumption for less critical services, optimizing the balance between reliability and complexity.
Solution Approach 2:
The patent uses configurable parameters in replication policies to control the degree of service replication and pre-configuration. By adjusting parameters such as the number of standby nodes, their readiness level, and resource allocation, the system can flexibly balance reliability requirements against complexity and resource constraints without requiring a complete redesign of the replication architecture.
3Speed
If multiple standby nodes are prepared with service instances, then failover speed is improved, but resource consumption and device complexity increase
Solution Approach 1:
The patent applies partial action by preparing only the necessary portion of standby nodes with service instances based on failover speed requirements and resource constraints. Instead of fully configuring all possible standby nodes, the system configures enough nodes to achieve the required failover speed while leaving other nodes in a less prepared state, thereby reducing resource consumption while maintaining adequate failover performance.
Data Source
AI summary
Techniques are disclosed for deploying services in a server cluster environment. Certain techniques are disclosed for deploying services to a cluster based on a replication policy that includes a plurality of configurable parameters. In some embodiments, the configurable parameters (also referred to herein as replication factors) can define a number of nodes to which a service is to be deployed, a number of nodes to which a service is to be prepared, and/or a number of nodes to which a service is replicated. Based on the configurable parameters, the replication policy enables users and/or cluster providers to guarantee different levels of performance and/or reliability.